[CliffordK (OP)]

There has been a recent discussion about thermal gradients and power generation.

Looking at thermoclines.
http://www.nodc.noaa.gov/cgi-bin/OC5/WOA09F/woa09f.pl?parameter=t

The water between Japan and Korea has a most interesting thermocline.  At the surface the water is around 10°C.  At about 200m, the temperature of the water (saltwater) drops down to near 0°C.

So, if one had a substance with a strongly negative thermal expansion coefficient through that range, it should naturally create strong convection currents. 

Pure water might fit the bill, with negative thermal expansion from about 4°C to 0°C.

D₂O has a higher freezing/melting point of about 3.8°C.
D₂¹⁸O would have even a higher freezing point.

So, one could either just create convection currents with pure water.
Or a column of pure "heavy water" should actually freeze at about 200m, with the heavy ice then floating to the surface (in the heavy water column) with significant force.

Would one want ice which might require an exotic water such as D₂O, HDO, or D₂¹⁸O?  Or would one just want to create the convection currents using liquid water?  One might also be able to mix substances to alter the negative thermal expansion range and freezing point of water, but I would worry that a mixture would tend to stratify over the depth.  But perhaps the convection currents would prevent stratification.

I suppose the last question is that if one was to choose to use ocean energy, is it appropriate to use an energy source that could disrupt the thermocline?  Or, would it be better to utilize deep sea currents, wave, and tide generators that might have less of an environmental impact?

[syhprum]

Power stations using the temperature gradient in the sea are certainly used in Japan there is certainly no need to use extremly exspencive Hydrogen isotopes.

[Geezer]

Power stations using the temperature gradient in the sea are certainly used in Japan

I remember seeing systems proposed that would do that, but I didn't realize anyone had actually done it. Do you have any more info?

[syhprum]

http://en.wikipedia.org/wiki/Ocean_thermal_energy_conversion

This type of power plant has a long history.

[CliffordK (OP)]

The plants may have a long history, but not a wide implementation.

If you look at the Japanese plant from the Wikipedia article above:

on the island of Nauru.  The plant became operational 1981-10-14, producing about 120 kW of electricity; 90 kW was used to power the plant and the remaining electricity was used to power a school and other places

So, the plant is about 25% efficient.  30KW is not much bigger than a person can get with a small portable gas/electric generator.  For comparison, our local landfill gas generator is a small plant with several 800KW Caterpillar engine/generators, and about 2MW total energy production.  We have a couple of very small hydroelectric plants on the McKenzie river producing 16.7MW and 8MW.  A larger dam in Washington produces 6GW of power.  So, the Japanese 30KW net generation is awfully small.

My proposal, using Negative Thermal Expansion (NTE) substances as part of the generation cycle might not be universally applicable, but for plants with NTE fluids, one could eliminate all the waste energy in pumping water from the deep, and in fact, likely generate more energy from the convection currents than would otherwise be realized from the temperature difference.  Ice would certainly give stronger currents, but might be difficult to work with.

It might be possible to do other mixes to slightly alter the freezing point of water, as well as the NTE temperature range of water.  Or, as mentioned, with the right temperature range, it might be possible to use even normal freshwater. 

Keep in mind, the competing technology, many nuke plants, already use D₂O in their cooling systems, and use very complex refining processes to make their fuel.

If the demand for D₂O, or even D₂¹⁸O was high enough, the production cost would go down.

I suppose the question should be how much energy it takes to pump a fluid to a depth of 500M with a positive thermal expansion, or pump water from the deep to the surface.  I'll try to work out the energy equations shortly.  Likewise, what would be the energy differential for plain water in the optimum temperature range of 4°C to 0°C?  What about using Ice for NTE?

[damocles]

Pure light water: H₂O

Density @ 10°C = 999.73 g/L
Density max @ 3.98°C = 1000.00 g/L
Density @ 0°C = 999.87 g/L
Ice @ 0°C = 916.8 g/L
Surface ocean water @ 10°C = approx 1023 g/L

Pure heavy water: D₂O

Melting point = 3.8°C
Density @ 3.8°C = approx 1105 g/L
Ice density @ 3.8°C = approx 1019 g/L
Density max @ 11.6°C = approx 1106 g/L

Conclusion: an impractically small amount of kinetic energy is available from dynamic exploitation of negative expansion temperature gradients unless ice is involved. Moving solid would pose some interesting design problems.
A pure water system would need isolation from sea water. Mixing of sea salt into the pure water would become a rogue issue, most particularly as salt concentration varies with depth in a way that varies with locality and season, and it is likely that negative thermal expansion could not be relied on for oceanic brines.
There are easier ways to earn a megawatt!

[CliffordK (OP)]

Thanks, Damocles,

That is just what I calculated (with water).

I looked up the difference of density due to thermal expansion.

Density of ice (pure water) at 0°C: 915 kg/m³
Density of pure water at 0°C: 999.9  kg/m³
Density of pure water at 4°C:1000 kg/m³

So, if one had a 1m² x 1000m column at 0°C, it would weigh 999,900 kg.
If one had a 1m² x 1000m column at 4°C, it would weigh 1,000,000 kg.
With the difference of 100kg, or about 10cm head.  Not nearly as much as I had hoped.

If, on the other hand, it was a column of pure ice, it would weigh 915,000 kg, or a difference of 85,000 kg, or about 85m of head, which wouldn't be bad.   That would, of course, also depend on the actual density of the ice and water at depth.

Perhaps one could make an ice slurry for the upward path.  But, overall, the pressure differences aren't as much as I hoped.  It also means that the pumping of the water would not be as energy  intensive as I had expected, although the energy gain would still be low.